Environmental Engineering Reference
In-Depth Information
(a) Different spatial shapes have to be linked depending on the character of the
input data and models used (e.g. joining point data, areal data and images)
(b) Considering different temporal and spatial situations means not simply asses-
sing data but interpolating them on a consistent temporal and spatial level (e.g.
calculating mean annual area-data from daily data of a station)
(c) Analyse the joint spatial information, characterizing its distribution and typify-
ing it (e.g. generating spatial zones of equal character)
(d) Finding appropriate indicators of the status of the individual system, undertak-
ing generalizations, dealing with the complexity of the information (e.g. calcu-
lating sum parameters)
All methods mentioned may include inaccuracy or fuzzy information. Uncertain-
ties as well as underlying assumptions ask for transparency, clearly reproducing the
process of integration, highlighting the methods used and the paradigms followed.
Additionally, visualization, presenting the results of an integrated model is the basis
of all discussions of the significance of the results (Turk 1992).
In the following the above mentioned methods are explained in detail:
a) Spatial Joining
The simplest way of
joining
data in a GIS is to upload geo-referenced files and
combine them in a map. There are two common kinds of geo-reference: coordinate
systems, defining position on the globe, and map projections, describing the
position on a plane. To couple data from different systems and projections (and
there are a lot of them) detailed metadata on methods, dates and sizes are needed to
transform a data set from one coordinate system to another. Metadata are “data
about data”, used to describe the definition, structure and administration of all
contents of a data file in its context. Finally, combining geo-files that establish
and maintain metadata is a basic requirement.
For
joining
non-georeferenced data, GIS offer two kinds of functions: Either, by
connecting data by known identifiers like attributes of places (or names of locations
as towns or water bodies), or by an ID or name of e.g. a sample station. Rules are
offered to either join objects that intersect each other, or where one object
completely falls inside another one. Both are feasible techniques for coupling
data and models. The result of joining attributes or locations adds to the input layer.
Joining data into a GIS by their coordinates is a common procedure. Assuming
that the position of a station is given as geographic longitude and latitude in
degrees, minutes and seconds, this information can be used to add an x-y-feature
in a GIS. As a rule, GIS uses decimal degrees, so minutes and seconds have to be
converted into decimal degrees. To fit the points from a coordinate system (e.g.
survey data from the investigation area) to a given layer in the GIS that is available
as a map projection, the positions have to be transformed. The coordinates of the
points are projected to the map, considering details like date, zero meridian, and
